Project Summary/Abstract African American (AA) women have a 40% higher breast cancer mortality than U.S. women of European ancestry (EA). This disparity reflects a number of underlying factors, including differences in clinical presentation [e.g., later stage at diagnosis and higher incidence of more aggressive tumors such as estrogen receptor negative (ER-) tumors] and distribution of established risk factors. One of the strongest and most well- established risk factors for breast cancer in EA populations is high mammographic density (MD), which refers to the proportion of breast tissue comprised of fibroglandular tissue (vs. adipose tissue) and can be readily measured on screening mammograms. Yet, there is very limited knowledge about the determinants of high MD or associations of high MD with breast cancer risk in AA women. Of the three studies to date, each with <200 cases in AA women, only one found a statistically significant association. To address these gaps in knowledge, we propose to establish a mammogram repository within the Black Women?s Health Study (BWHS), an ongoing cohort study of 59,000 AA women with over 20 years of comprehensive epidemiological and clinical data. Multiple screening (pre-diagnostic) full-field digital mammogram images will be retrieved for 6700 women, including at least 700 incident breast cancer cases. This resource will constitute the largest screening mammogram repository among AA women in the U.S. with individual-level risk factor data spanning 20 years. Percent and absolute MD will be quantitatively measured on all mammogram images using an established and validated interactive thresholding technique. The overarching goals of this research are to understand associations of MD with breast cancer in AA women and to identify predictors of high MD in this population. Relative risks will be precisely estimated for associations of high MD with breast cancer overall in AA women and, for the first time, by ER subtype. AA women have been greatly underrepresented in research on MD to date. This population-based research overcomes methodological limitations of the few previous studies and will generate new knowledge that is critical and necessary to reduce racial disparities in breast cancer. The large sample size ensures sufficient statistical power for carrying out the Aims proposed here as well as for providing a database for future research questions that will no doubt emerge with advances in the field of digital mammography imaging. In summary, this innovative research has great potential to advance current knowledge about breast cancer etiology in AA women, to inform opportunities for risk reduction or prevention, and to close the gap in racial disparities in breast cancer. In particular, information gained from the proposed research will be useful for risk prediction models for breast cancer in AA women.